HPC Challenge Benchmark
|Original author(s)||Innovative Computing Laboratory, University of Tennessee|
The HPC Challenge Benchmark is a set of benchmarks targeting to test multiple attributes that can contribute substantially to the real-world performance of HPC systems, co-sponsored by the DARPA High Productivity Computing Systems program, the United States Department of Energy and the National Science Foundation.
It consists at this time of 7 benchmarks: HPL, STREAM, RandomAccess, PTRANS, FFTE, DGEMM and b_eff Latency/Bandwidth. HPL is the LINPACK TPP (Toward Peak Performance) benchmark. The test stresses the floating point performance of a system. STREAM is a benchmark that measures sustainable memory bandwidth. RandomAccess measures the rate of random updates of memory. PTRANS measures the rate of transfer for larges arrays of data from multiprocessor's memory. Latency/Bandwidth measures latency and bandwidth of communication patterns of increasing complexity between as many nodes as is time-wise feasible.
The annual HPC Challenge Award Competition at the Supercomputing Conference focuses on four of the most challenging benchmarks in the suite:
- Global HPL
- Global RandomAccess (OR BSS Random Access Benchmark)
- EP STREAM (Triad) per system
- Global FFT
There are two classes of awards:
- Class 1: Best performance on a base or optimized run submitted to the HPC Challenge website.
- Class 2: Most "elegant" implementation of four or more of the HPC Challenge benchmarks.
- "Cray X1 Supercomputer Has Highest Reported Scores on Government-Sponsored HPC Challenge Benchmark Tests". 2004-06-14. Retrieved 2010-01-22.
- "HPCC FAQ". University of Tennessee. Retrieved 2010-01-22.
- "HPC Challenge Award Competition". DARPA HPCS Program. Retrieved 2010-01-23.